559 / 670
Table 2
Nominal
T
stat
p
-value
Mean
SD
T
stat
=
where: = mean of the differences between measured results and reference value
sd = standard deviation of the differences between results and reference value
n
= number of replicate analyses of SRM
H
0
: = 0, The null hypothesis is that there is no difference between the measured results and the certified value:
p
-value: is the probability (at the 95% levels of confidence) of obtaining the test statistic assuming that the null hypothesis is true.
p
-value < 0.05 indicate a difference of statistical signifance between measured results and SRM.
95% confidence interval: The true population of the measured value expected to fall within the confidence interval estimate 95% of the time.
lower limit = mean -
upper limit = mean +
where: = critical values of the t-distribution
sd = standard deviation of the differences between results an reference value
n
= number of replicate analyses of SRM
Bias
Accuracy - Analytical bias compared to reference material
Certified Reference Material
Measured Results
CRM product
Analyte
Certified value Uncertainty Certified range Concentration
Difference from
certified value
Confidence interval
n
/
sd
d
d
d
n
t
crit
sd
n
t
crit
sd
crit
t
SPIFAN ERP Checklist v1.0
Fos-01 (February 2016) - MEF
FOR ERP USE ONLY
DO NOT DISTRIBUTE